See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/315710705

CFD modeling for thermal performance of Closed Loop Pulsating heat pipe in
bottom heated mode

Conference Paper · February 2017

DOI: 10.15224/978-1-63248-118-4-33

CITATIONS READS

2 518

2 authors:

Ashutosh Gupta Ajit Parwani

Institute of Infrastructure, Technology, Research and Management Institute of Infrastructure Technology Research and Management, Ahmedabad
4 PUBLICATIONS   2 CITATIONS    12 PUBLICATIONS   46 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

inverse heat transfer View project

feasibility of carbon capture with CaL in India View project

All content following this page was uploaded by Ashutosh Gupta on 31 March 2017.

The user has requested enhancement of the downloaded file.

 Proc. of Sixth International Conference On Advances in Civil, Structural and Mechanical Engineering -ACSM 2017
Copyright © Institute of Research Engineers and Doctors, USA .All rights reserved.
ISBN: 978-1-63248-118-4 doi: 10.15224/ 978-1-63248-118-4-33
CFD modeling for thermal performance of Closed
Loop Pulsating heat pipe in bottom heated mode
Ashutosh Gupta, Ajit K Parwani
Abstract— In this research work a computational model for
closed loop pulsating heat pipe (CLPHP) has been developed in
order to study its two phase flow process and predict its
thermal performance. The pulsating heat pipe of 1.3mm
internal diameter in vertical bottom heat mode was simulated
with unsteady model. Water was taken as working fluid.
Volume of fluid model (VOF) in FLUENT was selected for
multiphase modelling, also continuum surface force (CSF)
model was considered for effects of surface tension. The
simulation with unsteady model was successful in reproducing
the two phase flow process in CLPHP and oscillation caused by
the pressure difference along with the startup characteristics.
During the working of CLPHP, lowest thermal resistance of
0.295 K/W was achieved. The simulation result obtained is
consistent with the experimental results at the same condition.
Keywords—Closed Loop Pulsating heat pipe,
Computational modelling, bottom heat mode, unsteady model,
two phase flow
I. Introduction
In order to maintain operational stability of electrical and
electronics components, heat or thermal energy must be
efficiently transferred to the surrounding. A heat pipe is
working on a two phase system has better heat transport
capability than a conventional single phase heat exchanger.
However, it has various limitations [1], e.g. capillary limit,
boiling limit, entrainment limit, sonic limit and viscous
limit. Some of these limitations will be overcome by
Pulsating heat pipe (PHP) which was introduced by Akachi
[2] in the middle of 1990s.
PHP has many advantages over conventional heat pipe,
for example it doesn‟t have wick structure, its
miniaturization can be easily made, easy to manufacture and
better heat transfer ability. Due to its advantages it has been
widely studied and applied in many areas such as thermal
management of electronic component, hybrid vehicle, fuel
cell cooling, integral thermal spreader, and wire-on-tube
heat exchanger [3-6].
The PHP consists of tubes/channels arranged in a
serpentine manner and joined end to end. The diameter of
the PHP tubes is of capillary diameter so that vapor plugs
can be formed due to capillary action. If the diameter is too
large, the liquid and vapor phases will stratify. The PHP can
operate successfully for all operation modes [6]. Generally,
there are two types of PHPs: the looped PHP, which has two
open ends, connected to one another, and the unlooped PHP,
which has two unconnected ends. Oscillating motions of
liquid plugs and vapor slugs in capillary diameter tube, drive
the working fluid flowing in the pipe. The liquid plugs and
Institute of Infrastructure Technology Research and Management
(IITRAM)
India
ajitkumar.parwani@iitram.ac.in
46
vapor slugs were formed in PHP and gets randomly
distributed in the pipe due to surface tension. The oscillation
motion counteracts the pressure difference due to expansion
and contraction of working fluid caused by absorption and
dissipation of sensible heat. The operational mechanism of
the PHP involves evaporation and condensation of thin
liquid film, dynamic contact angle of the working fluid on
tube wall, the growing and flowing of Taylor bubble and the
coalescence of the small bubbles [7]. Therefore, although
the structure of a PHP is very simple, the coupling of
hydrodynamic and thermo- dynamic effects during the
process of heat and mass transfer in the PHP make the
operational mechanism of PHP very complicated and hard to
be fully revealed [8,9].
Extensive experimental and theoretical work have been
conducted by many researchers over decades, studying
complex characteristics of heat transfer and dynamic
oscillating behavior inside PHP. In spite of its simple
structure, the PHP working principles are very complex and
yet a complete and validated design tool has not been
developed [8].
For experimental research, various parameters such as
geometric parameters of PHP, inner diameter, the cross-
section shape of the PHP, channel configuration, the length
of the evaporation and condensation section, the number of
turns, etc. have been intensely studied. Various operational
parameters such as charge ratio, heat flux, inclination angle,
external force, etc. have also been researched [8].
For theoretical studies, many models describing
dynamics inside CLPHP have been proposed. Wong et al.
[10] used Lagrangian approach for theoretical modeling of
PHPs in which they assumed the flow was under adiabatic
conditions for the entire PHP. For the vapor plugs they have
shown the pressure and velocity variations with time. Shafii
and Faghri [11] used a U-shaped model of open PHP for the
analysis of oscillating heat transfer. They developed
conservation equations for each liquid and vapor slug by the
control volume method. Their model was able to estimate
slug flow direction and the ratio of sensible and latent heat.
Mameli [12], considered mass, momentum and energy
conservation using finite difference scheme. The results
showed that heat transfer in both looped and unloop1ed
PHPs is due mainly to the exchange of sensible heat. Givler
and Martinez [13] took the homogeneous bubble model
(HBM) for PHP simulation by FLOW-3D. They showed
that most of the coalescing has taken place near the
condenser, and liquid slugs agglomerated according to the
direction of gravity. Lin [14] used commercial software
fluent to consider the effects of surface tension and phase
change on thermal performance of PHP. Nagwase [15] used
star CCM Plus to model two phase flow in PHP with DI
water as working fluid without considering temperature
dependency of different parameters of working fluid.
 Proc. of Sixth International Conference On Advances in Civil, Structural and Mechanical Engineering -ACSM 2017
Copyright © Institute of Research Engineers and Doctors, USA .All rights reserved.
ISBN: 978-1-63248-118-4 doi: 10.15224/ 978-1-63248-118-4-33
Though various numerical models have been proposed
over last two decades, but only a few of them were able to
obtain proper simulation of different PHPs. For example,
Zuo et al. [16] gave one dimensional model for the thermal
hydraulic simulation of a pulsating heat pipe and
experimental validation was shown only for the thermal
performance over filling ratio for a flat-plate PHP with
embedded porous wick. Khandekar and Groll [17]
suggested a lumped parameters model for the single loop of
CLPHP. He assumed homogeneous fog flow which seems
more suitable for thermosyphons than for the PHP. Hence
overall still there is no effective numerical model to predict
thermal performance of PHP.
To further explore the potential applications of PHPs
technology, it is necessary to predict the thermal
performance of CLPHPs by computational modeling and
simulation. Hence in this work a computational model for
CLPHP is developed to simulate the process of flow and
heat transfer, including the oscillations and the temperature
change due to evaporation and condensation occurring in
bottom heating mode. Experimental results [18] were
eventually used to verify the model established in this paper
II. Computational Modelling of
CLPHP
A. Physical Model
A two dimensional physical model was developed using
the commercial software package ANSYS FLUENT, to
simulate the internal flow and heat transfer in a CLPHP.
Water was used as working fluid. The copper was chosen as
tube material. According to the experimental conditions [18]
four turn PHP is divided into 3 parts; evaporator section,
adiabatic section, and condenser section. The corresponding
sections are shown in Fig 1. The geometric dimension of
CLPHP are mentioned in Table 1.
Fig 1. Geometric model
47
Table 1: Geometric dimension of physical model
Parameter Value (mm)
Inner diameter 1.3
Outer diameter 2.5
Evaporator length 20
Condenser length 20
Adiabatic length 150
The grid size was chosen according to Lin [14] with cell
count equal to 8400 where minimum and maximum length
of the element is taken equal to 0.026 mm and 2.6 mm
respectively. Time step of 10-5 is chosen.
Fig 2. Mesh configuration of evaporator section
Figure 2 shows meshing configuration used in this study.
Only a part of evaporator section has been shown in the
figure for clear visualization of quadrilateral mesh used in
the simulation.
B. Computational Model Set-up
Continuity equations, momentum equation and energy
equations in VOF model were used to describe the motion of
working fluid in a CLPHP [14]. Volume of fluid (VOF)
model with two Eulerian phases, water and water-vapour
with implicit body force formulation and 10 -8 volume
fraction cutoff was activated. Energy equation was turned on
to account for heat transfer. Viscous model was selected
laminar. Water vapor and water were taken as primary fluid
and secondary fluid respectively. The density (ρ) of water
was taken to be temperature dependent and its value was
calculated from steam table:
ρ = 80106.84 - 747.9185T + 2.35502T2 – 0.00247T3 (1)
Gravitational acceleration was turned on and operating
pressure of 4246.9 Pa was considered. As the CLPHP being
a negative pressure system, it was assumed that the boiling
temperature of water was 303 K, according to the
experimental observations. The phase change between liquid
and vapor is carried out using evaporation-condensation
model governed by Lee model [19]. The continuum surface
force (CSF) model was used to consider the effect of surface
tension. The value of surface tension was taken to 0.0712
N/m along with wall adhesion.
 Proc. of Sixth International Conference On Advances in Civil, Structural and Mechanical Engineering -ACSM 2017
Copyright © Institute of Research Engineers and Doctors, USA .All rights reserved.
ISBN: 978-1-63248-118-4 doi: 10.15224/ 978-1-63248-118-4-33

The heat flux in evaporation section depended on the power

input. According to the general temperature requirement of
electronic cooling and the experimental results [18], the
heating powers are not over 100 W. Hence maximum heat
flux of 100W is assigned at the evaporator. The convection
heat transfer coefficient in condensation section was set to
about 4000 W/ (m2 K). The adiabatic section was set as
insulation boundary condition. The convection heat transfer
coefficient for the other section which are in contact with air
was set to 5W/m2 K.

C. Solution Methods
The operating temperature was 301 K and the initialized
filling ratio was taken as 50% [18]. In the model solution the
pressure velocity coupling was taken as PISO with
skewness- neighbor coupling. Under discretization scheme,
pressure interpolation was set to body force weighted;
momentum, volume fraction and energy were taken as first
order upwind. The relaxation factors for pressure, density
and momentum were set to 0.3, 1, and 0.01 respectively.
These control factors however were varied during the
simulation to maintain the residuals in desired range. The
flow characteristics and thermal resistance for CLPHP were
obtained for time-step of 10-5 s.

III. Results and discussion

The simulation was carried out for a time period of 13s
in order to obtain two phase flow process occurring in Fig. 4 Contours of temperature change in CLPHP operation process
CLPHP and to determine its thermal resistance.. at 1.5s, 3.5s, 5s and 10s

There was a temperature-rise period before the CLPHP

developed into stable oscillation, which is clearly visible in

Fig. 3 Temperature overshoot resembling bubble growth at 1.5s

the Fig. 3. This operating temperature overshoot seen at 1.5s

of startup, represented the bubble generation in CLPHP, and
corresponding vapour bubble generated is also observed.
These observations clearly resembles startup performance of
bubble generation in CLPHP.
Temperature plots of CLPHP with water as working
fluid at different time can be seen in Fig. 4. Two phase flow
process and pulsation of liquid plugs due to pressure
difference between the evaporator and condenser was also Fig. 5 Volume fraction of water at 5s, 7.5s and 12s.

48
 Proc. of Sixth International Conference On Advances in Civil, Structural and Mechanical Engineering -ACSM 2017
Copyright © Institute of Research Engineers and Doctors, USA .All rights reserved.
ISBN: 978-1-63248-118-4 doi: 10.15224/ 978-1-63248-118-4-33
achieved as shown in the Fig. 5. It is found that pulsating
behavior of liquid plugs in CLPHP in vertical bottom heated
mode was attained after 7.5s of startup.
The thermal resistance (R) of CLPHP is the ratio of
temperature difference between the evaporator and
condenser and corresponding heating power. Hence
Te  Tc
R (2)
Q energy and the environment, China; 2003.
where Te is average temperature at evaporator; T c is average
temperature of condenser, and Q is the heating power.
Thermal resistance CLPHP with water as working fluid in
vertical bottom heated mode was plotted against simulation
time as shown in Fig. 6.
Fig 6. Graph of thermal resistance with time.
It can be seen from the figure that thermal resistance
becomes almost constant after 7.5s after which it can be
treated as normal pipe with constant thermal conductivity.
Hence it is also clear from the graph that steady pulsating
flow is achieved at 7.5s. Also because of temperature
overshoot at 1.5s highest thermal resistance of 0.56 K/W
was achieved.
From the computational model developed lowest thermal
resistance achieved was 0.295 K/W whereas thermal
resistance achieved in experimental work by Zirong Lin [15]
was found to be 0.24 K/W. Hence the simulations results are
in good agreement with respect to experimental work.
Conclusions
IV.
In this paper a computational model was developed to
understand the heat transfer mechanism in CLPHP and to
predict the thermal performance of CLPHP in vertical
position with water as a working fluid. The effects of phase
change using Lee model and surface tension using
continuum surface model was considered. Following
conclusion have been made from the study:
 It was shown that the simulation of CLPHP with
unsteady model was able to predict the operation of
CLPHP, including bubble formation and oscillation
phenomena caused by the pressure difference.
 Overall the computational model shows good
agreement with experimental work. Therefore, it can
serve as a design tool for estimating the thermal
performance of pulsating heat pipe with different
geometries and working fluid, without the need of any
experimental model.
49
References
[1] H.H. Yang, S. Khandekar, M. Groll, “Operational limit of
closed loop pulsating heat pipes,” Appl. Therm. Eng., vol. 28,
pp. 49–59, 2008.
[2] Akachi, H., Structure of a Heat Pipe, US patent, No. 4,921,041,
1990.
[3] Groll M, Khandekar S. Pulsating heat pipes: progress and
prospects. In: Proceedings of international conference on
[4] G. Burban, V. Ayel, A. Alexandre, Experimental investigation
of a pulsating heat pipe for hybrid vehicle applications, Appl.
Therm. Eng. 50 (1) (2013) 94–103.
[5] J. Clement, X. Wang, Experimental investigation of pulsating
heat pipe performance with regard to fuel cell cooling
application, Appl. Therm. Eng. 50 (1) (2013) 268–274.
[6] H. Yang, S. Khandekar, M. Groll, Performance characteristics
of pulsating heat pipes as integral thermal spreaders, Int. J.
Therm. Sci. 48 (4) (2009) 815–824.
[7] Khandekar S, Panigrahi PK, Lefèvre F, Bonjour J. Local
hydrodynamics of flow in a pulsating heat pipe: a review. Front
Heat Pipes 2010; (1):1–20.
[8] X. Han, X Wang, H Zheng. Review of the development of
pulsating heat pipe for heat dissipation. Renewable and
sustainable energy reviews, pp. 692-709.
[9] Zhang Y W, Faghri A. Advances and unsolved issues in
pulsating heat pipes. Heat Transf Eng 2008; 29(1): 20–4
[10] Wong, T. N., Tong, B. Y., Lim, S. M., and Ooi, K. T., 1999,
„„Theoretical Modeling of Pulsating Heat Pipe,‟‟ Proceedings
of 11th International Heat Pipe Conference, Tokyo, Japan, pp.
159–163.
[11] Y.W. Zhang, A. Faghri, M.B. Shafii, Analysis of liquid–vapor
pulsating flow in a U-shaped miniature tube, Int. J. Heat Mass
Transf. 45 (2002) 2501–2508.
[12] A. Faghri, Y. Zhang, Thermal modeling of unlooped and
looped pulsating heat pipes, J. Heat Transfer 123 (2001) 1159–
1172.
[13] R.C. Givler, M.J. Martinez, Modeling of Pulsating Heat Pipes,
SANDIA REPORT, SAND2009-4520, Springer, August 2009.
[14] Z. Lin, S. Wang, R. Shirakashi, Simulation of a miniature
oscillating heat pipe in bottom heating mode using CFD with
unsteady modeling, Int. J. Heat Mass Transfer 57 (2) (2012)
642–656
[15] Nagwase S Y, Pachghare P R. Experimental and CFD analysis
of closed loop pulsating heat pipe with DI-water. In:
Proceedings of ICEETS, New York; 2013.
[16] Zuo, Z. J., North, M. T., and Ray L., Combined Pulsating and
Capillary Heat Pipe Mechanism for Cooling of High Heat Flux
Electronic, Internal Report, Thermacore, Inc., Lancaster, PA,
USA, 2001.
[17] Khandekar S., and Groll, M., Roadmap to Realistic Modeling
of Closed Loop Pulsating Heat Pipes, Proc. 9th International
Heat Pipe Symposium, Kuala Lumpur, Malaysia, November
2008.
[18] Z.R. Lin, S.F. Wang, J.J. Chen, et al., Experimental study on
effective range of miniature oscillating heat pipes, Appl.
Therm. Eng. 31 (5) (2011) 880–886.
[19] Fadhl B, Wrobel LC, Jouhara H (2015) CFD Modelling of a
two-phase closed thermosyphon charged with R134a and
R404a. Appl Therm Eng (78). pp.482-490
 Proc. of Sixth International Conference On Advances in Civil, Structural and Mechanical Engineering -ACSM 2017
Copyright © Institute of Research Engineers and Doctors, USA .All rights reserved.
ISBN: 978-1-63248-118-4 doi: 10.15224/ 978-1-63248-118-4-33

About Author (s):

Ashutosh Gupta is B. Tech

IIIrd year student in
Mechanical Engineering
Department at IITRAM
Ahmedabad, Gujarat
(India). His main research
areas are in the field of heat
transfer.

Ajit K Parwani is Assistant

Professor in Mechanical
Engineering Department at
IITRAM Ahmedabad,
Gujarat (India). He graduated
from RGPV Bhopal and post
graduated from DTU Delhi,
and obtained his PhD from
IIT Delhi (India), in 2013. He
has been serving as an
academician and researcher
in technical institutes from 10
years in India. His main
research areas are in the field
of heat transfer, combustion
and renewable energy.

50

View publication stats